The enterprise AI land grab is on. Glean is building the layer beneath the interface.

The competition in the enterprise AI sector is becoming increasingly fierce. Major players like Microsoft are integrating Copilot into their Office suite, while Google is embedding Gemini into Workspace. Meanwhile, OpenAI and Anthropic are engaging directly with businesses, and virtually every SaaS provider is now offering an AI assistant as part of their package. Amidst this rush to enhance user interfaces, Glean is focusing on a less visible yet crucial aspect: creating a foundational intelligence layer that operates beneath these interfaces. Founded seven years ago, Glean originally aimed to be the go-to search engine for enterprises, offering an AI-driven tool that indexes and searches through various SaaS applications such as Slack, Jira, Google Drive, and Salesforce. However, the company has since pivoted from merely improving enterprise chatbots to establishing itself as the connective tissue between AI models and enterprise systems. In a recent discussion with TechCrunch at the Web Summit Qatar, Glean's co-founder Jain emphasized the importance of understanding user behavior and preferences in developing high-quality agents. While acknowledging the power of large language models (LLMs), he pointed out their limitations: they lack specific knowledge about individual businesses, including their workforce and operations. As a result, it is essential to merge the generative capabilities of these models with the unique context of each enterprise. Glean claims to have already mapped this context and acts as an intermediary between AI models and enterprise data. The Glean Assistant serves as an entry point for customers, offering a user-friendly chat interface powered by a combination of proprietary models like ChatGPT, Gemini, and Claude, along with open-source alternatives, all tailored to leverage the company’s internal data. What keeps clients engaged, according to Jain, is the robust infrastructure that supports these interactions. Glean provides model access that allows companies to utilize multiple LLMs without being tied to a single provider, facilitating adaptability as technologies advance. Jain sees major players like OpenAI and Google not as competitors but as collaborators in innovation. Glean also excels in integrating with key enterprise systems such as Slack, Jira, Salesforce, and Google Drive, helping to streamline information flow and enabling agents to operate within these tools effectively. Perhaps most critically, Glean has developed a governance framework that ensures data security and compliance. This framework allows the system to provide relevant information while considering user permissions, a vital feature for large organizations. As enterprises explore AI solutions, Jain warns against the pitfalls of loading all internal data into a model without a structured approach. It’s essential for Glean to ensure that the models do not produce erroneous outputs. Their system includes mechanisms to validate the accuracy of model responses against original documents, generate citations, and respect user access rights. The overarching question remains whether a dedicated intelligence layer can thrive as tech giants like Microsoft and Google deepen their foothold in the enterprise market. With both companies vying for increased control over enterprise workflows, it’s uncertain if standalone solutions like Glean will maintain relevance. However, Jain believes that companies prefer not to be locked into a singular AI model or productivity suite, indicating a demand for a neutral infrastructure layer that can coexist with larger platforms. Investors are backing this vision, as evidenced by Glean's successful $150 million Series F funding round, which nearly doubled its valuation to $7.2 billion. Unlike many AI startups that require substantial computing resources, Glean is focused on sustaining a healthy and rapidly growing business.